PROJECT SUMMARY/ABSTRACT Type 1 diabetes (T1D) is a complex disease that affects 1.25 million individuals in the United States and is characterized by autoimmune destruction of insulin-producing beta cells in the pancreatic islets. T1D GWAS studies have identified 114 index SNPs mapping to 58 loci that influence disease risk, but how these loci contribute to T1D pathogenesis is largely unknown. A major challenge in functionally characterizing T1D loci is identifying the precise causal variant(s) underlying disease risk at each locus. As the majority of T1D-risk variants are non-coding, an additional challenge is inferring the target genes of T1D-risk signals. Finally, functional validation of causal T1D-risk variants requires the use of appropriate experimental models for the key pathogenic cell types. To address these challenges we have assembled a team of highly accomplished investigators in human genome sciences (Gaulton and Frazer), epigenomics and high-throughput molecular assays (Ren), as well as T1D biology and diabetes models of human pluripotent stem cells (hPSC) (Sander). We propose to combine state-of-the-field computational methods, high-throughput molecular assays, and hPSC-based cell models to comprehensively identify T1D causal variants, annotate these variants across T1D- relevant cell types, and define their regulatory functions and target genes. In Aim 1 we will utilize T1D GWAS data, epigenomic annotation and high-throughput transcription factor binding assays to fine-map known T1D loci and identify and fine-map novel loci. These studies will annotate variants across cell types and thus be a broadly informative resource for the T1D community for hypothesis generation of variant function. In preliminary studies we discovered that ~50% of independent T1D-risk signals map to regulatory elements active in islets or pancreatic progenitors, suggesting a possible role of these variants in beta cell and progenitor function. In Aim 2 we will utilize a collection of 100 whole-genome-sequenced hPSCs, including 24 hPSC lines from T1D patients, to derive pancreatic progenitors and then generate Hi-C chromatin conformation, RNA-seq, ATAC-seq, H3K27ac ChIP-seq, and DNA methylation profiles. We will utilize these data combined with similar available datasets from human islets to annotate molecular QTLs at T1D-risk variants, as well as to evaluate potential contributions of rare variants to disease etiology in the 24 T1D patients. These analyses will provide insight into how T1D-risk variants alter local (<10kb) chromatin states and gene regulation in islets and their progenitors. In Aim 3 we will further uncover variant/target gene relationships by integrating the genetic data from Aim 1 and the local QTL data from Aim 2 with 3D chromatin contact maps to identify T1D-risk variants in enhancers that are distal (>10kb) QTLs for promoter sites. To validate predicted T1D enhancer-target gene(s) interactions and potentially discover new target genes, we will prioritize ten enhancers with strong molecular phenotypes for functional analysis by CRISPR/Cas9-mediated deletion in hPSCs. The proposed studies will provide key insight into the molecular and cellular functions of T1D-risk variants identified through GWAS.